0 penilaian0% menganggap dokumen ini bermanfaat (0 suara)
158 tayangan17 halaman
Luis andres bredlow explores some grammatical and logical aspects of "is" in Parmenides. He argues that parmenides' "is" is to be taken most plausibly as a copula of defnitional identity. This account will permit us to overcome some central difculties inherent in other predicative interpretations of "is," he says.
Luis andres bredlow explores some grammatical and logical aspects of "is" in Parmenides. He argues that parmenides' "is" is to be taken most plausibly as a copula of defnitional identity. This account will permit us to overcome some central difculties inherent in other predicative interpretations of "is," he says.
Hak Cipta:
Attribution Non-Commercial (BY-NC)
Format Tersedia
Unduh sebagai PDF, TXT atau baca online dari Scribd
Luis andres bredlow explores some grammatical and logical aspects of "is" in Parmenides. He argues that parmenides' "is" is to be taken most plausibly as a copula of defnitional identity. This account will permit us to overcome some central difculties inherent in other predicative interpretations of "is," he says.
Hak Cipta:
Attribution Non-Commercial (BY-NC)
Format Tersedia
Unduh sebagai PDF, TXT atau baca online dari Scribd
Reviewed work(s): Source: Classical Philology, Vol. 106, No. 4 (October 2011), pp. 283-298 Published by: The University of Chicago Press Stable URL: http://www.jstor.org/stable/10.1086/662036 . Accessed: 01/02/2012 06:48 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to Classical Philology. http://www.jstor.org 283 Classical Philology 106 (2011): 28398 [ 2011 by The University of Chicago. All rights reserved] 0009-837X/11/10604-0001$10.00 PARMENIDES AND THE GRAMMAR OF BEING luis andrs bredlow T HE AIM OF THIS PAPER is to explore some grammatical and logical aspects of the word is () in the fragments of Parmenides. I will argue that Parmenides is is to be taken most plausibly, in its frst and most immediate sense, as a copula of defnitional identity, expressing the essence or nature of something. This defnitional use implies both the absolute and the veridical sense of is. This account will permit us to overcome some central difculties inherent in other predicative interpretations of Parmenides is, such as those proposed by Alexander Mourelatos, Richard Ketchum, and Patricia Curd. 1 At the beginning of her discourse, the goddess points out the ways of inquiry that alone can be conceived (B2.38): , ( ), ,
( ), . Any interpretation of this crucial passage has to begin by answering two related questions: (1) what is the subject of is at B2.3 and 5? and (2) what is the sense of the word is? As for question (1), there is widely shared consensus in recent scholarship that we are not meant to supply any implicit subject of is, such as reality, truth, and so on. 2 This seems to leave us with the following alternative: (1a) we might understand the routes of is and is not as not representing actual propositions or theses, but rather types of proposition, propositional functions or sentence frames 3 (x is F, x exists, etc.); or (1b) we might take is as implying its own subject (and perhaps its own predicate). On the latter reading, is would be understood as saying something like what-is 1. For the interpretation of Parmenides is I will propose, I am indebted to some suggestions of Agustn Garca Calvo (1981, 1996). I am also indebted to the two anonymous referees of CP for helpful comments and criticism on earlier versions of this paper, and to Elizabeth Asmis for her detailed suggestions for stylistic improvement. 2. A recent exception is Gemelli Marciano (2009, 65, 68, 88), who holds that the subject of is is all things (B8.38, where we are given the solution of the riddle). 3. Mourelatos 1979, 3; cf. Mourelatos 1970, 5160, for a more detailed account. This line of interpretation goes back to Calogero (1932, chap. 1). 284 LUIS ANDRS BREDLOW is, what-there-is is what-there-is. 4 The solution I will argue for in this paper could be seen as a synthesis of both possibilities: Parmenides argument sets out to establish the logical conditions of what it means to be something, and ends up leading us to identify what-is or what-there-is (without any further qualifcation) as the only legitimate subject of is. For question (2) there is a widely divergent range of answers available in the literature: is has been interpreted as existential (x exists), veridical (It is the case that p), predicative (x is F ), or a fusion of these senses. At any rate, it seems clear that the rejection of the way of is not is argued for at B2 on the grounds that one cannot know what-is-not, nor express it or talk of it (or show the way to it or point it out, depending on how one prefers to translate ). On the existential interpretation, this argument becomes utterly unsound: 5 of course we can talk of the nonexistent, 6 maybe even know it. We can know, to begin with, that it does not exist, which may well be the only thing worth knowing about it. On occasion, this may even be an important insight: one may fnd out, for example, that there is no void, no life on Mars, or no rational square root of 2. Therefore, to rule out statements of inexistence as illegitimate would betray a rather awkward understanding of scientifc inquiry, even for an Archaic thinker. Moreover, it is at least doubtful that , even in later philosophical use, should be understood as equivalent to the verb exist in modern languages. For example, as Charles H. Kahn and Uvo Hlscher have convincingly shown, none of the four meanings of discussed by Aristotle at Metaphysics 5.7 can be identifed as properly existential in the modern sense, that is, as neatly detached from the copula or predicative use. 7 So it seems quite safe to conclude, with Kahn, that the Greeks did not have our notion of existence 8
(even though Kahns own veridical interpretation of Parmenides is as primarily referring to a Wittgensteinean Bestehen von Sachverhalten 9
seems hardly less anachronistic 10 ). 4. See Bollack 1965, 3: 143 (cf. 2006, 20, 11316); Cassin 1980, 55; Garca Calvo 1981, 176; Cordero 1984, 77, and 2004, 6064; Finkelberg 1988, 45. 5. So do, in fact, almost all of Parmenides arguments. As Ketchum (1990, 186) rightly states, Most of what Parmenides tells us about why what-is-not is unthinkable, etc., is false and, if I may say so, fairly obvi- ously false on the existence interpretation. 6. We can, of course: mermaids, for instance, argues Owen (1960, 94 n. 1). 7. Kahn 1966, 24849; Hlscher 1976, 1331. On the absence of a neat distinction between existential and predicative uses of is in Aristotle, see also Hintikka 1986; Brown 1994; and Bck 2000; on the absence of a notion of existence in the modern sense in ancient Greek, cf. Graham 1965, 22425; Heitsch 1974, 10611; and Kahn 1976 and 1979. 8. Kahn 1966, 248; cf. Kahn 1973 and 1976. 9. Kahn 1966, 251, 262; cf. Kahn 1968, 71113. 10. Against Kahns interpretation, Mourelatos (1968, 742) points out that the Greeks conceived of the world as the totality of things and not of facts, as Wittgenstein did in the Tractatus; similar objections were raised by Hlscher (1976, 48) and Klowski (1977, 111). For more detailed criticism of Kahns theory of the veridical use of Greek is, see Benardete 1977 and Matthen 1983. It should be noted, however, that in his more recent work Kahn (2002, 2004, 2005) has modifed his views considerably. Nevertheless, he is still con- vinced that the veridical sense fts best with Parmenides claim that you cannot know what-is-not (fr. 2,7), although other values of einai are also required for his argument (Kahn 2004, 391 n. 14). But on this interpre- tation, as Kahn has to acknowledge, Parmenides argument still ends up in fallacy and confusion (Kahn 2002, 90; cf. below, n. 16). 285 PARMENIDES AND THE GRAMMAR OF BEING Consequently there has been an increasing number of recent scholars who prefer to understand Parmenides is in a mainly predicative sense, that is, as functioning like the copula in propositions of the form x is F. 11 I think this goes in the right direction; however, some qualifcations seem necessary. In the frst place, a predicative interpretation should not be taken to imply that the predicative sense of is is the only one that should be taken into consideration. A predicative interpretation, as I see it, is compatible with a fusedor rather, undiferentiatednotion of being, covering various uses that modern logic distinguishes from one another, including the predicative, veridical, and absolute or existential uses. 12 This seems indeed to be the most natural understanding of Parmenides is, the one that comes closest to current uses of / both in ordinary Greek language and in philoso- phy. However, this assumption leaves still open the question how exactly these various senses of is are related to one another. For example, is the fused sense to be taken as a mere juxtaposition, or rather as an organically articulated structure, within which some specifc use(s) of is may prove to be more fundamental than others? At any rate, a fusion of these various senses of is (preserving logical consistency) should be distinguished from their simple confusion (which would make Parmenides reasoning depend on some fallacy of equivocation). 13 Second, and more importantly, the predicative reading of is is in itself open to diferent interpretations. Ketchum, for example, suggests that Par- menides to be, in its relevant syntactically incomplete occurrences, is to be taken as meaning to be something or other. 14 Hence its negation, not to be, would be equivalent to not to be anything at all. This seems to make quite good sense of the rejection of is not at B2: whatever we are inquiring into either is something or other, or it is not anything at all. The second term of the alternative is rightfully ruled out by the goddess as a path entirely unlearnable, since what is not anything at all cannot be known or described or pointed out. 15 On this interpretation, however, as Ketchum himself acknowledges, the argument begins to lose its cogency at B8.67. In these verses, Parmenides argues that what-is cannot have come to be from what is not, for it is not sayable or thinkable that it is not. Parmenides assumption is that, if what-is has come to be, then there must have been a time at which it was not, which 11. See Mourelatos 1970 and 1975; Austin 1986; Ketchum 1990; Curd 1998b, 3463. Earlier proponents of a predicative reading of Parmenides is were Hofmann 1925, 8; Cassirer 1925, 41; and Calogero 1932. 12. A fused sense of isthough with a strong focus on the existential componentwas suggested by Furth (1974 [originally published 1968]); cf. also Furley 1973, 1314, and Goldin 1993, 20 n. 1. A fused or rather undiferentiatedmeaning of is centered on the predicational use has been most explicitly ad- vocated, though not developed in detail, by Kerferd 1972. For Hintikka (1980, 6), the decisive feature of Parmenides conceptual framework is the absence of any distinction between what we twentieth-century philosophers would take to be diferent senses of is . . . , viz. the is of existence, the is of predication, and the is of identity. 13. This sort of fallacy (confusion of the predicative and the existential is) was attributed to Parmenides by Calogero 1932, followed by, among others, Kirk and Raven 1957, 270; but this charge is generally rejected in more recent literature (see Mourelatos 1979, 3). 14. Ketchum 1990, 16768. 15. Ketchum 1990, 17375. 286 LUIS ANDRS BREDLOW according to B2 is impossible. Stated this way, however, the argument is ambiguous. At frst sight, we might understand that, if what is something or other has come to be a thing that is something or other, then there must have been a time at which it was nothing at all. This would be true enough; but it does not warrant Parmenides conclusion that what-is must be eternal and un- generated. There still remains the possibility that what is something or other (for example, F) might have come to be what it is now, F, after having been something diferent (for example, G). There would be no need of there having been any time at which it was not anything at all. In order to conclude that what-is is ungenerated, Parmenides would need a diferent kind of argument: he would have to assume that, if what is F has come to be F, then there must have been a time at which it was not anything at all. Now this is clearly a non sequitur. We would have to conclude that Parmenides, caught up in the traps of his language, failed to see that a thing can fail to be something or other (to have a particular predicate) without being nothing at all (or failing to have any predicate at all). In other words, he confused is not something (is not F ) with is not anything. 16 This seems rather unsatisfactory. Of course, we cannot exclude a priori the possibility that some of Parmenides reasoning might be fallacious after all. Nonetheless, I think we should not be ready to resign ourselves to this conclusion unless we have made sure there is no consistentand at the same time historically and philologically plausiblereconstruction of his argument available. A proposal that comes quite close to meeting this triple requirement is the one ofered by Curd in her Legacy of Parmenides, elaborating on some earlier suggestions by Mourelatos. Curd takes Parmenides is to be an informative identity claim, an assertion that, when true, reveals the nature of a thing, saying just what something is, insofar as what we know in knowing what-is is the real or genuine character of a thing. 17 This interpretation places Parmenides in his proper Greek context, since it clearly connects his notion of Being with the Ionian quest for the physis or nature of things. Moreover, it permits us to make reasonable sense of his argument against coming-to-be: Whatever something genuinely is cannot be something that becomes; it must always remain what it is. 18 Thus Par- menides argument against coming-to-be (as well as the rest of his arguments in B8, as Curd has quite accurately shown) is consistent if we take it as refer- ring specifcally to predications concerning the genuine nature or essence of things. I think this view is basically right. However, Mourelatos (1999) has pointed out two central difculties inherent to Curds account (and to his own, as he readily acknowledges). In the frst place, both interpretations seem to require rather severe restric- tions on the range of possible subject and/or predicate terms for is. On 16. Ketchum 1990, 18082. Similarly, Kahn (2002, 90) holds that for Parmenides, [d]epending on which function of einai is being denied, to m on can represent either negative predication, falsehood, non-identity, non-existence, or non-entity, that is to say, nothing at all. The fallacy in Parmenides argument lies . . . in the confused union of these various modes of negation in the single conception of what-is-not. 17. Curd 1998b, 39. 18. Curd 1998b, 77. In a very similar sense, Mourelatos (1970, 107) had already stated: What a thing gets to be (or comes to be) is not what the thing really is, i.e. in its essence or nature. 287 PARMENIDES AND THE GRAMMAR OF BEING Mourelatos account, the x of the Parmenidean x is F scheme ranges over any and all ordinary physical objects, whereas the F ranges over the physis or altheia (nature, reality, or true identity) of the ordinary object at issue, like in claims such as Wool is compacted air, This world is ever- living fre, and so on. 19 On Curds interpretation, it is already the subject term of isthe x of the x is F schemethat is to be seen as restricted to a quite special kind of entities: Parmenides subject is an entity of a certain kind, one that plays a fundamental role in an explanation of what there is. In other words, the subject of is can only be something that would count as a basic entity in an explanation of how things are. 20 So it seems, at any rate, that in order to make sense of Parmenides argu- ments we have to assume that these are meant to be understood as being true only if referring to a rather special kind of subject or predicate. But since both the subject and the predicate terms of is are elided or merely implicit, such a proposal is indeed open to the objection of specious complexity and artif- ciality, especially if this restriction is taken to afect not only the predicate, but also the subject of is. 21 The second difculty concerns the motive for the rejection of the route of is not. For Curd, as for Mourelatos, this motive is to be found in the vagueness inherent in any negative statement about the nature or essence of things: For what would it be to give a negative account of what something really is? It would be to say it is not this, or this, or this . . . and such a route can never be completed; it is not anyston. Something that is negative in this sense would be nothing in particular, and it is clear that one cannot think of something that is nothing in particular. 22 In a similar vein, Mourelatos had put the question: Where do I go if I am told to go to not-Ithaca? . . . It is fated to be a journey of , of regress and wandering. In a word, the route of is not is a route of vagueness in the ancient and literal sense of vagrancy and divagation. 23 But to this line of argument, as Mourelatos himself acknowledges, one might all too easily object that negative statements need not be always vague, indeterminate, or uninformative. Some negative statements may have a quite determinate meaning, especially if the negated predicate term bears a relation of complementarity (or enantiomorphy, as Curd terms it) to some other well-defned positive term: for example, to be up just means to be not- down, and vice versa; to be light means to be not-dark, and vice versa. Such a relation of complementary opposition is precisely the one that exists between the two elements of the erroneous worldview of mortals in the Doxa: Night is precisely not-Light, and vice versa. So if Parmenides ruled out 19. Mourelatos 1999, 120. 20. Curd 1998b, 39 n. 42, 40 n. 46. 21. Mourelatos 1999, 121. Similarly, Graham (2006, 158) observes that on the predicative view Parmenides argument at B2.68 assumes a great deal of prior understanding on the part of the reader. Graham considers this a decisive objection to the predicative reading. 22. Curd 1998b, 49; cf. 61: To say that X is really not-F, or to attribute to X a nature or character that contains negative elements is to make a statement that is steeped in vagueness, which gives one nothing on which to pin ones thought. 23. Mourelatos 1970, 78. 288 LUIS ANDRS BREDLOW this conception on the ground that negative statements are necessarily vague and indeterminate, we would have to conclude that he committed the er- ror of assimilating the conceptually and semantically well-charted privation involved in enantiomorphy with the bald and inchoate is really not-F of the second route of inquiry cited in B2. 24 It should be noted that this second difculty is closely related to the frst. The exclusion of negative statements as vague or uninformative only seems plausible if referring to a rather special kind of subject. As Mourelatos points out, on his view the rejection of the negative route at B2 is not a rejection of negative predication in general, but rather a rejection of negative attributes in answer to speculative, cosmological questions. 25 Moreover, a further restric- tion has to be put on possible predicate terms. A statement of the form x is not F is indeed incurably vague if F is a proper name or an individual term: to know that something is not Ithaca or not Socrates hardly qualifes as knowing what this thing really is. But it will be much less vague if F is a com- mon term defned within a genus or family (for example, not blue is natu- rally understood as meaning any other color but blue). And it is clearly not vague if F is one of a pair of complementary or enantiomorphic terms, such as up/down, light/darkness, and so on. Consequently Mourelatos was drawn to stipulate, in his Route of Parmenides, that the predicate terms of the x is F scheme have to be of the categorial rank of individuals. 26
This seems rather gratuitous. In his later writings, Mourelatos insists that we can make sense of Parmenides argument against negative statements only if we suppose him committed to a naive metaphysics of things. Such metaphysics, according to Mourelatos, envisaged the world as consisting of distinct self-contained entities, each of which can be grasped by a mere act of acquaintance. Parmenides, lacking the conception of a genuine predicate, was unaware of the diference between predicates and proper names. 27 In the end, this is precisely the kind of solution one would have rather preferred to avoid: that is, the all-too-familiar expedient of saddling our philosopher with some kind of naive or primitive logic or metaphysics in order to excuse the apparent faws of his argument. I think, however, that there is a way of understanding Parmenides is that allows us to overcome both difculties. It is not very far from those proposed by Mourelatos and Curd, and compatible with what is most valuable and interesting in their contributions. To begin with the frst problem: the need for artifcial restrictions to the range of possible subject or predicate terms for is could evidently be avoided if it could be shown that the restrictions needed to make sense of Parmenides arguments are already implicit in the logical functioning of the word is itself, such as Parmenides means us to understand it. 24. Mourelatos 1999, 124. 25. Mourelatos 1970, 75. 26. Mourelatos 1970, 79. 27. Mourelatos 1973, 19, 43; cf. Mourelatos 1975 and 1999, 122, 124. 289 PARMENIDES AND THE GRAMMAR OF BEING I suggest that this is precisely what we obtain if taking Parmenides is as what Kahn calls a verb of whatness or the is of defnitional identity. 28
The is of whatness expresses what something is by defnition, its essence or nature. It is the is we fnd in the Socratic question , in Platos () , and Aristotles and , but also in the Ionian inquiry into the physis or nature of things. The most characteristic trait of defnitional predication, as I understand it, is the implication that the predicate expresses something inherent in the genuine nature or character of something or somebody. It means that the predicate belongs to its subject essentially, constitutively, and inseparably. What a def- nitional predication says is that x is F in such a way that x cannot ever fail to be F under any circumstances. This comes quite close to what in modern philosophy is called an analytical proposition, but with at least one important diference: a defnitional predication, in the Greek sense, is not true by virtue of meaning alone. It is only true insofar as it refers to some real thing and expresses the genuine nature of this thing. The truth of such a predication is not assured by the meaning of words, as with the analytical propositions of modern philosophy, but by the very nature of things themselves. This approach allows us some useful specifcations concerning the fused or undiferentiated sense of is. In the frst place, the defnitional interpreta- tion does not exclude the absolute sense of , habitually translated by there is, 29 but rather implies it: only what there is can have an essence or nature. Indeed, it seems unlikely that anyone who was not acquainted with the refnements of modern (or, for that matter, medieval) logic would have suspected that there could be an essence or nature, or even a true defnition, of something that there is not or that does not exist. 30 It should be noted that the inverse implication probably does not obtain: the examples of Anaximanders , Platos , or the Stoic concept of suggest that Greek philoso- phers could quite well conceive of there being something that had no defnite or defnable essence at all. Very much of the same goes for the veridical sense of is: to defne the essence or nature of something is to defne what this thing truly or really is. To put it the other way round: to know the truth about something is to know its essence or nature, to know what this thing really is. So Parmenides is indeed concerned with the truth of statements, though of a quite peculiar kind: state- ments expressing essence or defnitional identity. In a word, the defnitional or whatness sense of is implies both the absolute and the veridical sense, but is not implied by either of these. 31 28. Kahn 1966, 26263; 1979, 29 n. 13. 29. This absolute sense is not exactly the same as the supposed existential sense of is: see Kahn 1966, esp. 24748; cf. also Ketchum 1990, 170 n. 6; Sedley 1999, 115. 30. According to Aristotle (An. post. 2.7.92b48), there can be defnitions only of really existing things. The medieval notion of an essence (essentia, quiditas) entirely independent from actual existence (as in, e.g., Thomas of Aquinas, De ente et essentia, 4.97102) is clearly contrary to Aristotles own thought. 31. It might be remarked that these implications hold already for the predicative or copula use of is in Greek in general, of which the defnitional or whatness use is a special case. A predicative assertion (x is F ) normally implies the existence of its subject, as well as the truth of what it asserts. Inversely, existence or truth claims regularly imply some predicative proposition (see Kahn 2004), though not necessarily a predica- tion of defnitional identity. 290 LUIS ANDRS BREDLOW That Parmenides means us to understand his is in precisely this defni- tional sense is indicated already by the modal qualifer contained in the crucial assertion It is and cannot not be (B2.3). If this is meant to be one statement and not two, 32 we probably have to understand it as: It is F in such a way that it is not possible for it not to be F. This is most prominently the case of defnitional predication, since it is obviously impossible for something not to be what it is by defnition. In terms of Parmenidean imagery, defnition is the limit (B8.26, 31, 42) in whose bonds (B8.2631) what-is is held by the forceful necessity (B8.30) of being what it is. 33 As A. H. Coxon put it, is means that its subject is something, and that being this precludes it . . . from not being what it is and a fortiori from not being anything at all. 34 This elementary assumption of logic, if applied with sufcient intransi- gence, may well account for all of Parmenides theorems in B8. To begin with, if x is, by defnition, F (or F is the essence or nature of x), then x cannot have become F, nor ever cease to be F, for in either case it would be true at some moment that x is not F, which contradicts the hypothesis (cf. B8.521). The all-too-obvious objection to this argument is that x might have begun to exist, or may cease to exist altogether at some moment, without ever being x and not being F at the same time. This objection would probably be met by Parmenides arguing that, since x cannot have come to be out of nothing (B8.710), there must be some entity y which at some moment has become or ceases to bex and, therefore, F. Hence F would properly be a predicate of y rather than of x. But since y has become (or ceases to be) F, it is evident that F cannot be the essence or nature of y. Whatever other term we may choose, the argument will repeat itself until we arrive at some predicate term F such that x cannot ever have become F, nor cease to be F. For similar reasons, x cannot be more or less F in some of its parts than in others (B8.2225), since something either is or is not F, with no intermediate degrees (B8.11, 16). If x lacks anything to be F, it lacks everything, that is, it is not F (B8.3233). Even minimal change is excluded (B8.2631), probably on the ground that, once the possibility of change is admitted, any precise limit between essential and inessential changes would depend on arbitrary convention, and not on the nature of the thing itself. 35 To summarize, in the words of Curd, for Parmenides to be is to be the genuine nature of a thing, to be just what a thing is; to be such a nature, something must be what it is unchangingly, completely, and as a unity. 36 From this it follows that, inversely, the objects of ordinary reality, reputed by mortals to come to be, change, and perish, cannot properly be in the strict logical sense required by the goddess reasoning, but only by dint of human convention (B8.3841). 32. Note that the modal qualifer and cannot not be is omitted in the mentions of the frst way we fnd later on in the text ( , B8.2; , B8.16). 33. See Heitsch 1974, 175; Garca Calvo 1996, 14; cf. Tarn 1965, 117, 151. It seems signifcant that the technical terms for defnition used in later Greek philosophy (, , ) draw on the very same image of the limit which was frst employed by Parmenides (, B8.26, 31, 42, 49). 34. Coxon 1986, 175. 35. This is well illustrated by the Sorites paradox: see Wheeler 1983, 29394; cf. also Melissus B7.2. 36. Curd 1998a, 20. 291 PARMENIDES AND THE GRAMMAR OF BEING So the fact that is, in the Parmenidean sense, can be truly asserted only of some very special kind of entity (which must be eternal, unchanging, etc.) results as the logical outcome of the whole argument. There is no need to suppose that such a restriction is implicitly postulated by his reasoning from the outset. 37 It is more than probable, of course, that what Parmenides had in mind were just such theoretically fundamental entities as the of Ionian physics. But this supposition is by no means necessary to guarantee the consistency of his argument. The argument indeed works perfectly well without any a priori restrictions on possible subject or predicate terms for is, as long as is is consequently understood as a defnitional predication. We might even take the argument still one step further. Since no kind of transformation of one sort of stuf into another (for example, of air into stone or vice versa, as in Anaximenes) can be ruled out a priori, we have to conclude that the only legitimate predicateand hence the only legitimate subjectof is is what is or what there is in general: it is the subject-predicate that was already implied in the word is itself. Whatever there is can, in principle, become or cease to be this or that; but nothing can ever come to be out of nothing or dissolve into nothing. Therefore, the total mass of what-there-is remains always constant, eternally and unchangingly. The impossibility of not-being is the ultimate limit ( , B8.42) that guarantees the necessary self-identity of what-there-is. So paradoxically, the mere indeter- minate mass of what-there-is, without any further qualifcation, turns out to be the only real entity that truly meets the formal requirements for a perfect defnition: it is what it is, and cannot ever be anything else but what it is. One may, of course, postulate other entities that equally meet Parmenides requirements, being eternal, unchanging, homogeneous, and so on: Empedo- clean or Anaxagorean elements, Democritean atoms, Platonic forms, Aristo- telian species. However, the widely noted indebtedness of later Greek thought to the legacy of Parmenides should not blur one fundamental diference: all these basic entities of post-Parmenidean speculation are, at bottom, mere conjectures. None of them exhibits a self-evident logical necessity like the one that what-is is what it is. We may regret that Parmenides truth has little informative value; but as so often, a greater richness of content is bought at the price of a loss of apodictic certainty. So when we have arrived at the end of the way, we learn that we have returned to the starting point. 38 Now we see that the formula (it) is was already the whole truth, since all that can be truly said of what-is is: It is. 39
37. A similar conclusion has been reached, in diferent ways, by Sedley (1999, 114): As for why no sub- ject is made explicit [for is at B2], the safest answer is that at this stage we are still investigating the logical behaviour of the verb to be. Only in the light of this investigation will we be able to answer the question what can stand as the subject of is. Thus, identifying the proper subject of the verb to be is the fnal goal of the Way of Truth, not to be prejudged at the outset. 38. Cf. the announcement of the circular structure of the goddess reasoning at B5: And it is all one to me from where I should start; for I shall come back there again (Mourelatos translation). The interpreta- tion I propose might also provide an additional motive for preferring, at B1.29, Simplicius reading (well-rounded truth), rather than Sextus (well-persuasive truth); but noth- ing hinges on this controversial point. 39. This implication was clearly seen by Furth (1974, 264) and by some earlier interpreters of Parmenides: see Cassirer 1925, 41; Calogero 1932, 30; and Cherniss 1935, 383. 292 LUIS ANDRS BREDLOW All the rest are but explications of what was already implicit in this one word. Thanks to these explicationsthe signposts along the way of is (B8) we have learned to identify the only possible subject-predicate term of is: it is precisely the one already implicit in the defnitional copula itself. If this is could be (and probably had to be) understood at the outset as a mere sentence frame or propositional function, this form now has become, so to speak, flled with semantic content: it implies its own subject and its own predicate. Before going on, I have to deal with a possible objection. The is of whatness or defnitional predication might be suspected to be a highly sophis- ticated technical device of post-Socratic philosophy (like the notions of es- sence or defnition themselves), rather than a feature of ordinary language that Parmenides could expect to be transparent to his readers or listeners. It is not an accident that the examples of defnitional use of is given by Kahn are all Platonic or Aristotelian. Indeed, it seems impossible to fnd any strictly grammatical feature distinguishing this specifc sense from other uses of the copula. The grammatical clumsiness of Platos and Aristotles technical terms ( , , etc.) betrays the utter difculty of making their readers understand something that was clearly not immediately evident for ordinary speakers of Greek. Any reader of Plato recalls how Socrates interlocutors, when asked to give a defnition of what is just, pious, beautiful, and so on, are invariably quite at a loss to understand just what they are be- ing asked for, and only come up with random examples or enumerations. To understand what is meant by a defnition, as it seems, requires a good deal of philosophical trainingsurely more than could be expected from the public of a pre-Socratic writer. I have given this objection the most compelling formulation I could; but I think it can be met by starting from the spontaneous answers to Socrates questions we fnd in Platos dialogues. We may all too easily dismiss these wrong answers to the Socratic question What is . . . ? as inappropri- ate, naive, or even stupid. However, precisely these answers may give us a precious clue to the pre-philosophical, ordinary language understanding of defnitional predication. Despite their inadequacy from Socrates or Platos point of view, these answers characteristically meet what we have stated as the minimum requirement for defnitional predication: they state some neces- sary feature of the subject, something the subject cannot fail to be or to do. Euthyphro, for example, when asked for a defnition of piety, answers that it is pious to insist on the legal prosecution of wrongdoers, whoever they be and whatever their crime; not to do so is impious and ungodly (Euthphr. 5de). Though this is clearly not an adequate defnition of piety, the underlying idea is transparent: to prosecute wrongdoers in any given case is what we would call an essential or necessary feature of any pious person; it is something such a person cannot possibly fail to do. Typically, this kind of defnition takes on the form of an (interminable) enumeration like Someone who is x (pious, just, virtuous, brave, etc.) is somebody who, given a situation type y 1 , does z 1 (and cannot fail to do so), and, given a situation type y 2 , does z 2 , and . . . (see, for example, Menos list of virtues at Meno 71e72a). We may smile at this inability to understand Socrates questions; but this was, after all, how ordinary speakers of Greek used to answer this kind of question. 293 PARMENIDES AND THE GRAMMAR OF BEING Their failurefrom Platos point of viewis that, instead of trying to give a general defnition of some abstract notion or quality, they randomly name some feature (an ability, characteristic behavior, etc.) they consider essential to a thing or person who has this quality. (Their world, after all, is a world of concrete things and persons, not of abstract entities like justice, piety, beauty, etc.) The point here is that they all understand quite well that this feature must be an essential one, one that belongs necessarily to the character of this kind of thing or person. We might say: it is something such a thing or person must be or do by defnition. This aspect of necessity becomes patent when Socrates interlocutors are confronted with counterexamples proving that the feature in question is not necessarily inherent in this qual- ity: for example, that in some cases a just person will not give back what he has borrowed (Resp. 331cd). When faced with such counterexamples, they recognize immediately that the defnition was wrong, this time with no need for further instruction. 40 The same pattern of thought is prominent already in Homer, for example, in Odysseus reply to Circe (Od. 10.38385): , , , ; Circe, what man that is right-minded could bring himself to taste of food and drink, before he had yet won freedom for his comrades, and beheld them before his face? (Trans. Mur- ray [1919] 1984) A man who is right-minded, Odysseus argues, cannot fail to act in such and such a manner. The use of the copula implies here, once again, the im- possibility for its subject to behave otherwise; the connotation of necessity or constraint is evident. And this is precisely what I have called the pre- philosophical, ordinary language understanding of defnitional predication. It is a pattern of thought and language that was accepted and applied by ordinary speakers of Greek long before anybody had heard of any technical notion of defnition or essence. To say that something or somebody is this or that, in the defnitional sense, means that it is impossible for this person or thing not to be this under whatever circumstances. I think this was all Parmenides needed to get his argument started. For the rest, he just had to apply with intransigent rigor the normally established conventions concerning this kind of predication in order to show that the logical conditions implied by these conventions cannot truly obtain for any of the ordinary objects of reality, nor for the basic entities of earlier cosmological speculation. There is still another possible source of inspiration for the Parmenidean use of the defnitional copula, which has already been pointed out by a minor- ity of interpreters. 41 Parmenides might well have been acquainted with the 40. A similar point has been made by Alexander Nehamas, Confusing Universals and Particulars in Platos Early Dialogues (1999, 15975): what Socrates interlocutors typically come up with when trying to answer the What is . . . ? questions are not particular instances of the concept they are being asked to defne, but rather particular ways of being pious, just, courageous, etc., or kinds of piety, justice . . . (I owe this refer- ence to an anonymous referee of CP.) 41. See, e.g., Kneale 1961, 90, 98; Cerri 1999, 46. 294 LUIS ANDRS BREDLOW necessary truth of mathematical propositions, 42 which would have ofered him a clear paradigm of defnitional predication. A mathematical proposition says that x is F 43 in such a way that it cannot not be F at any time nor under any circumstances (nor can it be more or less F to any, even minimal, degree of variation). To be sure, there is no hint of mathematical knowledge in the extant fragments of Parmenides. However, Epicharmus (23B2 DK) well-known mockery of Eleatic reasoning seems to imply that numbers and magnitudes were a familiar paradigm case of the invariance and self-sameness of somethings being what it is, in the strong Parmenidean sense. This would not mean, of course, that Parmenides conceived of what-is as a purely abstract or ideal mathematical entity. 44 Rather he searched for some equivalent for such timeless and unchanging entities in the physical world, and found it in the notion of an unchanging totality of what-there-is. In other words, Parmenides quest for necessary truth can be seen as a search for a kind of statement about physical objects (about something there is) that has the same apodictic certainty as mathematical or geometrical propositions. The way to fnd out what kind of statement this may be, and what kind of object it might refer to, consists in working out the ontological consequences (the signposts along the way of is) implied by the logical form of this kind of proposition. Now I come to the second difculty: the purported vagueness of negative predications. The solution I want to propose is already implicit in the sug- gested interpretation of is as a copula of defnitional predication. On this interpretation, is means x is, by defnition or essentially, F. This implies that x is F is logically equivalent to the tautology x is x. Hence its nega- tion, is not, or x is not F, means that x is not what it is by defnition. It means, ultimately, that x is not x, which is clearly self-contradictory. If this view is right, the route of is not is not rejected because of the vagueness supposedly inherent in negative statements, but because of the contradiction involved in the negation of a defnitional predication. This pattern of argument may be discerned in Parmenides reasoning in B8. If you should choose to say that what-is has become what-is, you will be committed to the assertion that at some moment it was true that what-is is not what-is, that x is not x, which is self-contradictory (B8.20). So you will learn that you are on the wrong road, the unthinkable (B8.17) path of is not. This reconstruction may seem open to an objection that has been raised against other predicative interpretations of Parmenides is: 45 Parmenides is not concerned with a negative predication of the form x is not F (for F being 42. Not necessarily from his often-alleged contacts with Pythagoreanism (as surmised by Kneale [1961, 90]; cf. the critical remarks of Owen [1966, 31718]). Indeed, as Burkert (1972, 40182) has shown, the contribution of Pythagoreans to the rise of early Greek mathematics seems to have been hugely overrated in both ancient and modern historiography. But there was, on the other hand, a considerable development of mathematics and geometry in Greece during the sixth and ffth centuries B.C.E., outside and independent from the Pythagorean milieu (see Burkert 1972, 41527, for a concise and well-informed survey). 43. It might be remembered that this is, in Greek, the standard form of mathematical propositions such as , twice fve are [= is] ten (Xen. Mem. 4.4.7). 44. notion, I think, utterly alien to any pre-Platonic thinker: Zeno (29B5, A24 DK), as it seems, still took for granted that what is not in any place cannot be anything at all. 45. Against Mourelatos, by Furley (1973, 1213) and Graham (2006, 158). 295 PARMENIDES AND THE GRAMMAR OF BEING some particular predicate), but with one that implies that x is nothing at all (B6.2, B8.10). I think, however, that this objection can be met if we suppose that Parmenides assumption is that these two kinds of negation are equivalent in the case of defnitional predications. If x is not what it is by defnition, then x cannot be anything at all, and hence it is necessary for it not to be ( , B2.5). For example, if it can be shown that the moving object is at rest (cf. Zeno A27), then we will have to conclude that there is no such thing as a moving object. In other words, the term moving object lacks reference or means nothing. The same assumption seems to underly Parmenides reason- ing in B8: if at any given moment x was not F, then at that moment it cannot have been anything at all. So the negation of a defnitional predication x is F, reducible to x is x, clearly entails the negation of x is in the absolute sense: if x is not F, and hence x is not x, then x is nothing. So the two routes of inquiry of B2 (It is, and cannot not be, and It is not, and it is necessary for it not to be) form an exhaustive alternative, once we understand the argument as concerned with essential or defnitional predica- tion alone (where x is F is equivalent to x is x): either x is x, or x is not x, which is absurd. The other two modal forms of predication (x is F, but can be not-F , and x is not F, but can be F ) are intentionally left aside as irrelevant to the issue of essential or defnitional predication. We might also say, as John Palmer has recently put it, that Parmenides is concerned at B2 with necessary truth and its reverse, (logical) impossibility, or, more exactly, with necessary being and necessary non-being. Palmer adds that Parmenides conceives of these modalities as ways of being or ways an entity might be rather than as logical properties. 46 This qualifcation seems to be required since a necessary truth in the modern sensean analytical proposition 47 need not be a truth about a necessarily existing object. 48 But as I have pointed out, defnitional predication, in the Greek sense, difers from modern analytical propositions in at least one important aspect: it is true only if it refers to some real thing whose genuine nature it expresses. For the Greeks any predication of the form x is F (be it analytical in the modern sense or not) is true only if there really is or exists some x of which F is true. So it seems to follow that x is F can be necessarily true only if x necessarily exists. Therefore Parmenides was right, I think, when assuming that necessary truth must refer to a necessarily existing object. In addition, this interpretation explains an apparent inconsistency of the goddess wording. At B2.2 she presents the route of is not as one of the routes of inquiry that can be conceived ( ). Later on, however, she 46. Palmer 2008, 3.5; for a detailed development of this modal interpretation, see Palmer 2009, esp. 83105. 47. I refer to analytical propositions in the rather broad sense in which the term is standardly used by most contemporary philosophers. An anonymous referee of CP points out to me that Parmenides claims might well apply to the synthetic a priori as well. I think this is surely right: Parmenides conclusions in B8 may well be understood as synthetic a priori judgments in the Kantian sense, if we assume that Kants distinction between the analytic and the synthetic a priori makes sense at all (this is denied by, e.g., Bennett 1966, chap. 1). I think it does (for detailed argument, see Bredlow 2010, 2432), but I see no urgent need to burden my argument with a discussion of this controversial topic. 48. A similar point was made by Owen 1966, 317 (against Kneale 1961) for the relation between timeless statements and timeless objects. 296 LUIS ANDRS BREDLOW insists that is not cannot be conceived (B8.89), and the route of is not is explicitly marked as inconceivable (B8.16). This is easily explained if is not stands for self-contradiction and hence logical impossibility. We cannot conceive, of course, that x is not x; but we surely canand mustconceive the impossibility of x not being x. Evidently, the recognition that a statement is self-contradictory entails the certainty that this statement is false. Thus the route of is not is indeed in a certain sense a legitimate way of inquiry: in logical terms, it is the method of reductio ad absurdum. But it is a route that ends as soon as it begins: once it is recognized as such, there is nothing more to fnd out on this route. So there remains only one route to talk about, namely, that of it is (B8.12). 49 Universidad de Barcelona 49. Moreover, if this interpretation is right, another often-stated problem can be dispensed with as well: if Parmenides does not rule out negative predication as such, but only negations of defnitional predications (i.e., self-contradictory statements), then there will be no need to seek for justifcations for the abundance of negative predicates in his own arguments. LITERATURE CITED Austin, S. 1986. Parmenides: Being, Bounds, and Logic. New Haven, Conn. Bck, A. T. 2000. Aristotles Theory of Predication. Leiden. Benardete, S. 1977. The Grammar of Being. Review of Metaphysics 30: 48696. Bennett, J. 1966. Kants Analytic. Cambridge. Bollack, J. 1965. Empdocle. 3 vols. Paris. . 2006. Parmnide: De ltant au monde. Lagrasse. Bredlow, L. A. 2010. Kant esencial. Barcelona. Brown, L. 1994. The Verb To Be in Greek Philosophy: Some Remarks. In Language, ed. S. Everson, 21236, Companions to Ancient Thought 3. Cambridge. Burkert, W. 1972. Lore and Science in Ancient Pythagoreanism. Trans. E. L. Minar, Jr. Cam- bridge, Mass. Calogero, G. 1932. Studi sull eleatismo. Rome. Cassin, B. 1980. Si Parmnide. Paris. Cassirer, E. 1925. Die Philosophie der Griechen von den Anfngen bis Platon. In Lehrbuch der Philosophie, ed. M. Dessoir, 7139. Berlin. Cerri, G. 1999. Parmenide: Poema sulla natura. Milan. Cherniss, H. 1935. Aristotles Criticism of Presocratic Philosophy. Baltimore. Cordero, N.-L. 1984. Les deux chemins de Parmnide. 2nd ed. 1997. Paris. . 2004. By Being, It Is: The Thesis of Parmenides. Las Vegas. Coxon, A. H. 1986. The Fragments of Parmenides. Assen. Curd, P. 1998a. Eleatic Arguments. In Method in Ancient Philosophy, ed. J. Gentzler, 128. Oxford. . 1998b. The Legacy of Parmenides. Princeton, N.J. Finkelberg, A. 1988. Parmenides Foundation of the Way of Truth. In Oxford Studies in Ancient Philosophy, vol. 6, ed. J. Annas, 3967. Oxford. Furley, D. J. 1973. Notes on Parmenides. In Exegesis and Argument: Studies in Greek Philosophy Presented to Gregory Vlastos, ed. E. N. Lee, A. P. D. Mourelatos, and R. M. Rorty, 115, Phronesis Suppl. Vol. 1. Assen. 297 PARMENIDES AND THE GRAMMAR OF BEING Furth, M. 1974. Elements of Eleatic Ontology. In The Pre-Socratics: A Collection of Critical Essays, ed. A. P. D. Mourelatos, 24170. Garden City, N.Y. Originally published in JHPh 6 (1968): 11132. Garca Calvo, A. 1981. Lecturas presocrticas. Madrid. 3rd ed. Zamora, 2001. . 1996. Ser y Realidad. Mania 2: 915. Gemelli Marciano, M. L. 2009. Die Vorsokratiker. Vol. 2, Parmenides, Zenon, Empedokles. Dsseldorf. Goldin, O. 1993. Parmenides on Possibility and Thought. Apeiron 26: 1935. Graham, A. C. 1965. Being in Linguistics and Philosophy. Foundations of Language 1: 22331. Graham, D. W. 2006. Explaining the Cosmos: The Ionian Tradition of Scientifc Philosophy. Princeton, N.J. Heitsch, E. 1974. Parmenides: Die Anfnge der Ontologie, Logik und Naturwissenschaft. Munich. Hintikka, J. 1980. Parmenides Cogito Argument. Ancient Philosophy 1: 516. . 1986. The Varieties of Being in Aristotle. In The Logic of Being, ed. S. Knuutila and J. Hintikka, 81114. Dordrecht. Hofmann, E. 1925. Die Sprache und die archaische Logik. Tbingen. Hlscher, U. 1976. Der Sinn von Sein in der lteren griechischen Philosophie. Heidelberg. Kahn, C. H. 1966. The Greek Verb To Be and the Concept of Being. Foundations of Language 2: 24565. . 1968. The Thesis of Parmenides. Review of Metaphysics 22: 700724. . 1973. The Verb Be in Ancient Greek. Dordrecht. . 1976. Why Existence Does Not Emerge as a Distinct Concept in Greek Philosophy. Archiv fr Geschichte der Philosophie 58: 32334. . 1979. Linguistic Relativism and the Greek Project of Ontology. Neue Hefte fr Phi- losophie 1516: 2033. . 2002. Parmenides and Plato. In Presocratic Philosophy: Essays in Honour of Alex- ander Mourelatos, ed. V. Caston and D. Graham, 8193. Aldershot. . 2004. A Return to the Theory of the Verb Be and the Concept of Being. Ancient Philosophy 24: 381405. . 2005. Parmenides and Being. In Frhgriechisches Denken, ed. G. Rechenauer, 21726. Gttingen. Kerferd, G. B. 1972. Review of Mourelatos 1970. Archiv fr Geschichte der Philosophie 54: 8992. Ketchum, R. J. 1990. Parmenides on What There Is. Canadian Journal of Philosophy 20: 16790. Kirk, G. S., and J. E. Raven. 1957. The Presocratic Philosophers. Cambridge. Klowski, J. 1977. Parmenides Grundlegung seiner Seinslehre B27 (Diels-Kranz). RhM 120: 97137. Kneale, W. 1961. Time and Eternity in Theology. Proceedings of the Aristotelian Society 61: 87108. Matthen, M. 1983. Greek Ontology and the Is of Truth. Phronesis 28: 11335. Mourelatos, A. P. D. 1968. Comments on The Thesis of Parmenides. Review of Metaphysics 22: 73544. . 1970. The Route of Parmenides. London. . 1973. Heraclitus, Parmenides, and the Naive Metaphysics of Things. In Exegesis and Argument: Studies in Greek Philosophy Presented to Gregory Vlastos, ed. E. N. Lee, A. P. D. Mourelatos, and R. M. Rorty, 1648, Phronesis Suppl. Vol. 1. Assen. . 1975. Determinacy and Indeterminacy as the Key Contrast in Parmenides. Lampas 8: 33443. . 1979. Some Alternatives in Interpreting Parmenides. Monist 62: 314. . 1999. Parmenides and the Pluralists. Apeiron 32: 11729. Murray, A. T., trans. [1919] 1984. Homer: Odyssey. Vol. 1. Loeb Classical Library. Reprint. Cambridge, Mass. 298 LUIS ANDRS BREDLOW Nehamas, A. 1999. Virtues of Authenticity: Essays on Plato and Socrates. Princeton, N.J. Owen, G. E. L. 1960. Eleatic Questions. CQ 10: 84102. . 1966. Plato and Parmenides on the Timeless Present. Monist 50: 31740. Palmer, J. 2008. Parmenides. Stanford Encyclopedia of Philosophy, http://plato.stanford.edu/ entries/parmenides/. . 2009. Parmenides and Presocratic Philosophy. Oxford. Sedley, D. 1999. Parmenides and Melissus. In The Cambridge Companion to Early Greek Phi- losophy, ed. A. A. Long, 11333. Cambridge. Tarn, L. 1965. Parmenides. Princeton, N.J. Wheeler, S. C. 1983. Megarian Paradoxes as Eleatic Arguments. American Philosophical Quar- terly 20: 28795.
Lorin W. Anderson, David R. Krathwohl-A Taxonomy For Learning Teaching and Assessing - A Revision of Bloom's Taxonomy of Educational Objetives-Longman (2001)